Matrices are polymeric networks characterized by insolubility in water. One type of polymeric matrix is a hydrogel, which can be defined as a water-containing polymeric network. The polymers used to prepare hydrogels can be based on a variety of monomer types, such as those based on methacrylic and acrylic ester monomers, acrylamide (methacrylamide) monomers, and N-vinyl-2-pyrrolidone. To form the gel, these monomer classes are typically crosslinked with such crosslinking agents as ethylene dimethacrylate, N,N'-methylenediacrylamide, methylenebis(4-phenyl isocyanate), ethylene dimethacrylate, divinylbenzene, and allyl methacrylate.
Another type of polymeric network can be formed from more hydrophobic monomers and/or macromers. Matrices formed from these materials generally exclude water. Polymers used to prepare hydrophobic matrices can be based on a variety of monomer types such as alkyl acrylates and methacrylates, and polyester-forming monomers such as .epsilon.-caprolactone and lactide. When formulated for use in an aqueous environment, these materials do not need to be crosslinked, but they can be crosslinked with standard agents such as divinyl benzene. Hydrophobic matrices can also be formed from reactions of macromers bearing the appropriate reactive groups such as the reaction of diisocyanate macromers with dihydroxy macromers, and the reaction of diepoxy-containing macromers with dianhydride or diamine-containing macromers.
Although there exist a variety of methods for producing polymeric networks, when these networks are intended to be created in the presence of viable tissue, and/or to contain a bioactive compound, the number of acceptable methods of producing polymeric networks is extremely limited.
It is nevertheless desirable to form both hydrogel and non-hydrogel polymeric matrices in the presence of viable tissue or bioactive agents for the purposes of drug delivery, cellular immune isolation, prevention of post-surgical adhesions, tissue repair, and the like. These polymeric matrices can be divided into two categories: biodegradable or bioresorbable polymer networks and biostable polymer networks.
Biodegradable polymeric matrices have been previously suggested for a variety of purposes, including controlled release carriers, adhesives and sealers. When used as controlled release carriers, for instance, polymeric matrices can contain and release drugs or other therapeutic agents over time. Such matrices can be formed, for instance, by a number of different processes, including solvent casting hydrophobic polymers. Solvent casting, however, typically involves the use of organic solvents and/or high temperatures which can be detrimental to the activity of biological materials and can complicate production methods. Solvent casting of polymers out of solution also results in the formation of uncrosslinked matrices. Such matrices have less structure than crosslinked matrices and it is more difficult to control the release of bioactive agents from such matrices. Yet another process, which involves the polymerization of monomers in or around the desired materials, suffers from cytotoxicity of monomers, oxygen inhibition and heat of polymerization complications.
Another process used in the past to prepare biodegradable and biostable hydrogels involves the polymerization of preformed macromers using low molecular weight initiators. This process involves a number of drawbacks as well, however, including toxicity, efficacy, and solubility considerations. For instance, when using a macromer solution containing a low molecular weight soluble initiator to encapsulate viable cellular material, the initiator can penetrate the cellular membrane and diffuse into the cells. The presence of the initiator may involve some toxic consequence to the cells. When activated, however, these initiators produce free radicals having distinct cytotoxic potential. Other drawbacks arise if the initiator is able to diffuse out of the formed matrix, thereby producing toxicity and other issues. Such initiators also tend to aggregate in aqueous solution, causing efficiency and reproducibility problems. Finally, in view of the limited efficiency of many initiators for initiating the necessary radical chain polymerization, it is often necessary to add one or more monomeric polymerization "accelerators" to the polymerization mixture. Such accelerators tend to be small molecules capable of penetrating the cellular membrane, and often raise cytotoxic or carcinogenic concerns.
U.S. Pat. Nos. 5,410,016 (Hubbell, et al.) and 5,529,914 (Hubbell, et.al.) for instance, relate to hydrogels prepared from biodegradable and biostable polymerizable macromers. The hydrogels are prepared from these polymerizable macromers by the use of soluble, low molecular weight initiators. Such initiators can be combined with the macromers, and irradiated in the presence of cells, in order to form a gel that encapsulates the cells.
Hydrogels often suffer from similar or other drawbacks in use as biological adhesives or sealants, e.g., for use as tissue adhesives, endovascular paving, prevention of post-surgical adhesions, etc. In each of the applications, the hydrogel matrix must generally "adhere" to one or more tissue surfaces. Current methods rely upon physical "adhesion" or the tendency of hydrogels to "stick" to a surface. A superior adhesive would provide both physical and chemical adhesion to surfaces utilizing the same physical characteristics as current hydrogel adhesives, but also providing chemical, covalent coupling of the matrix material to the tissue surface. Covalent bonds are generally much stronger than physical adhesive forces, such as hydrogen bonding and van der Waals forces.
As described above, when various techniques are used to form polymeric matrices via photoinitiation of macromers, the photoinitiators utilized tend to be low molecular weight. Polymeric photoinitiators have been described as well, although for applications and systems quite distinct from those described above. See, for instance, "Radical Polymerization", C. H. Bamford, pp. 940-957 in Kroschwitz, ed., Concise Encyclopedia of Polymer Science and Engineering, 1990. In the subsection entitled "Photosensitized Initiation: Polymeric Photosensitizers and Photoinitiators", the author states that "[p]olymeric photosensitizers and photoinitiators have been described. Many of these are polymers based on benzophenone, e.g., poly(p-divinylbenzophenone) (DVBP). Such rigid polymers are reported to be effective sensitizers since hydrogen abstraction from the backbone by excited benzophenone is less likely."
U.S. Pat. No. 4,315,998 (Neckers) describes polymer-bound photosensitizing catalysts for use in the heterogeneous catalysis of photosensitized chemical reactions such as photo-oxidation, photodimerization, and photocyclo addition reactions. The polymer-bound photosensitizing catalysts are insoluble in water and common organic solvents, and therefore can be readily separated from the reaction medium and reaction products by simple filtration.
What is clearly needed are macromers and macromer systems that avoid the problems associated with conventional polymeric matrices, and in particular, those drawbacks that arise when polymeric matrices are formed in the presence of viable tissue or bioactive agents.